Skip to content

Synergy map R7: #402 (adder words, vectorized exp, Abakus cost gate) - #403

Merged
AdaWorldAPI merged 1 commit into
masterfrom
ccr-45a41fb4-0j3i0l
Oct 30, 2026
Merged

AdaWorldAPI merged 1 commit into
masterfrom
ccr-45a41fb4-0j3i0l

Conversation

@AdaWorldAPI

Copy link
Copy Markdown
Owner

Adds section R7 to .claude/knowledge/synergy-map-2026-10-29.md for #402, and records the rows #400 and #401 moved, since neither had an entry. The change is documentation only.

Rows that moved

New dots

  • N22: half_add_mask / full_add_mask are facade words, but no consumer outside the fixtures uses them yet. A carry-save count over many mask words is the obvious candidate.
  • N23: every exp-based word got faster at once. Against Tract's Erf the gap is now about 4×, down from about 10×, and Tract pays for that in accuracy.
  • N24: AVX2 select still packs its masks lane by lane (42 vpextrw/vpinsrw). That is a second array-backed AVX2 cost next to mask_blend.
  • N25: NEON simd_exp_f32 has about 47 scalar fcmp/csel, and these predate D-BINARY-ABACUS-POLYFILL-0 PR A: half/full adder mask words; vectorized pow2n_from_int #402.
  • N26: the eager domino counter is not transactional, and Abacus is. The cost is equal on dense masks, slower and noisier on sparse.

Fruit

id fruit size
L20 diagnose N25 low
L21 harvest lance-graph mask-risc for a consumer of the adders low
H12 AVX2 compare-mask packing for select high
H13 promote Abacus, blocked on L21 high

PR C (the facet bit-plane helper) is not planned.

The map also records that #402 merged while four test jobs were still running, so their outcome was not seen here.

The facet-stride guard is clean on the diff.


Generated by Claude Code

Records #402, plus the rows #400 and #401 moved. New dots N22-N26: the
adders as facade words with no consumer yet, the exp speed-up reaching
every exp-based word, AVX2 select mask packing as a second array-backed
cost, NEON scalar fcmp/csel in simd_exp_f32, and the transactional
Abacus against the non-transactional eager counter. Fruit L20, L21, H12,
H13; PR C not planned.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XAqKMPh2Seh2WLccCksAdY
@coderabbitai

coderabbitai Bot commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor

Important

  • 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

Run configuration
  • Source mode: Auto
  • Repository: Unavailable
  • Branch: Unavailable
  • File: No configuration file from this repository applied
  • Commit: Unavailable
  • Layers: Organization UI
  • Review profile: CHILL
  • Plan: Advanced
  • Run ID: 48d6afc8-92f4-4654-aa21-bf70f9a3eeba
  • Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

@AdaWorldAPI
AdaWorldAPI marked this pull request as ready for review October 30, 2026 12:23
@AdaWorldAPI
AdaWorldAPI merged commit 55ec81e into master Oct 30, 2026
21 checks passed
@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Oct 30, 2026 •

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-10-30T12:25:26.366560Z cea7994 Draft marked ready
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants